search

google-ai-edge / mediapipe

Cross-platform, customizable ML solutions for live and streaming media.
https://mediapipe.dev
Apache License 2.0
26.78k stars 5.09k forks source link

landmark detection model on custom data #4908

Open sowmyakavali opened 10 months ago

sowmyakavali commented 10 months ago

Have I written custom code (as opposed to using a stock example script provided in MediaPipe)

None

OS Platform and Distribution

windows 10

Python Version

3.11

MediaPipe Model Maker version

No response

Task name (e.g. Image classification, Gesture recognition etc.)

landmark detection

Describe the actual behavior

I think now the customization only works for object detection and classification

Describe the expected behaviour

Can I train pose landmark model on custom data?

Standalone code/steps you may have used to try to get what you need

I want to train mediapipe landmark model on custom data, how can I do it?

Other info / Complete Logs

No response

kuaashish commented 10 months ago

Hi @sowmyakavali,

Similar to issue #4907, as already communicated within that specific issue, our current re-trainable capabilities encompass Objection Detection, Image Classification, Gesture Recognition, Face Stylization, Image Generation, and Text Classification, as detailed in the documentation.

We value and acknowledge your interest in expanding these capabilities. We will categorize this as a feature request and share it with our team. The prioritization of this request will be determined through internal discussions aligning with team priorities. Be assured that we will keep you informed of any updates concerning the potential inclusion of this feature.

Thank you!

sowmyakavali commented 10 months ago

Hi @sowmyakavali,

Similar to issue #4907, as already communicated within that specific issue, our current re-trainable capabilities encompass Objection Detection, Image Classification, Gesture Recognition, Face Stylization, Image Generation, and Text Classification, as detailed in the documentation.

We value and acknowledge your interest in expanding these capabilities. We will categorize this as a feature request and share it with our team. The prioritization of this request will be determined through internal discussions aligning with team priorities. Be assured that we will keep you informed of any updates concerning the potential inclusion of this feature.

Thank you!

Thanks @kuaashish